metadata
license: apache-2.0
language:
- en
base_model:
- HuggingFaceTB/SmolLM2-135M-Instruct
pipeline_tag: text-generation
library_name: transformers
tags:
- text-generation-inference
SmolLM2-135M-Instruct-GGUF
SmolLM2-135M-Instruct : The 135M model was trained on 2 trillion tokens using a diverse dataset combination: FineWeb-Edu, DCLM, The Stack, along with new filtered datasets we curated and will release soon. We developed the instruct version through supervised fine-tuning (SFT) using a combination of public datasets and our own curated datasets. We then applied Direct Preference Optimization (DPO) using UltraFeedback.
Model Files
| File Name | Size | Format Description |
|---|---|---|
| SmolLM2-135M-Instruct.F32.gguf | 540 MB | Full precision (32-bit floating point) |
| SmolLM2-135M-Instruct.BF16.gguf | 271 MB | Brain floating point 16-bit |
| SmolLM2-135M-Instruct.F16.gguf | 271 MB | Half precision (16-bit floating point) |
| SmolLM2-135M-Instruct.Q8_0.gguf | 145 MB | 8-bit quantization |
| SmolLM2-135M-Instruct.Q6_K.gguf | 138 MB | 6-bit quantization (K-quant) |
| SmolLM2-135M-Instruct.Q5_K_M.gguf | 112 MB | 5-bit quantization (K-quant, medium) |
| SmolLM2-135M-Instruct.Q5_K_S.gguf | 110 MB | 5-bit quantization (K-quant, small) |
| SmolLM2-135M-Instruct.Q4_K_M.gguf | 105 MB | 4-bit quantization (K-quant, medium) |
| SmolLM2-135M-Instruct.Q4_K_S.gguf | 102 MB | 4-bit quantization (K-quant, small) |
| SmolLM2-135M-Instruct.Q3_K_L.gguf | 97.5 MB | 3-bit quantization (K-quant, large) |
| SmolLM2-135M-Instruct.Q3_K_M.gguf | 93.5 MB | 3-bit quantization (K-quant, medium) |
| SmolLM2-135M-Instruct.Q3_K_S.gguf | 88.2 MB | 3-bit quantization (K-quant, small) |
| SmolLM2-135M-Instruct.Q2_K.gguf | 88.2 MB | 2-bit quantization (K-quant) |
Quants Usage
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):
